The Prospect of Using Cyanobacterial Bicarbonate Transporters to Improve Leaf Photosynthesis in C3 Crop Plants

The photosynthetic CO2-fixing enzyme Rubisco arose some 3.5 billion years ago, in an environment when CO2 was high and oxygen (O2) was low. Under these conditions, it was CO2 saturated and presumably performed well (Badger et al., 1998). However, since the advent of oxygenic photosynthesis, the levels of O2 have risen dramatically and CO2 has fallen to very low levels. This has gradually created conditions where CO2 has become limiting for Rubisco and allowed O2 to act as an alternative inhibitory substrate for the enzyme. To cope with these dramatic environmental changes, two major strategies have evolved to help Rubisco maximize its carboxylation rate at ambient levels of limiting CO2. First, the enzyme has evolved better kinetic properties, where the K m(CO2) has decreased and the ability to distinguish against O2 has increased at the expense of catalytic rate (Badger et al., 1998). Alternatively, many photosynthetic organisms, ranging from cyanobacteria to algae to land plants, have developed active CO2-concentrating mechanisms (CCMs) to turbo-charge Rubisco’s CO2 supply at a minor metabolic cost (Badger et al., 1998). Most notably, among plants this has led to the development of C4 photosynthesis (Sage, 2004). Most of the important grain crops (rice [Oryza sativa], wheat [Triticum aestivum], barley [Hordeum vulgare], canola [Brassica napus], soybean [Glycine max]), tuber crops, and vegetable crops are C3 species and have applied the first strategy and lack any form of CCM at the leaf or chloroplast level. Much of the inherent inefficiency in C3 photosynthesis revolves around the need to gain CO2 through passive diffusion through the leaf pores (stomata), across cell walls and cytoplasm, and eventually through to the chloroplasts. Diffusive resistance to CO2 passage results in a drawdown of the effective CO2 concentration in the chloroplast, and C3 plants have adopted strategies to maximize the diffusive conductivity for CO2 by appressing chloroplasts against the intracellular airspaces and having large chloroplast surface area-to-leaf area ratios (Evans and von Caemmerer, 1996). Low chloroplast CO2 concentrations exacerbate the CO2 limitations and increase the wasteful Rubisco oxygenation reaction of ribulose 1,5-bisphosphate (RuBP) to produce phosphoglycolate, which must be recycled back to RuBP through a complex set of reactions known as the photorespiratory cycle. This is worsened by increased temperature, with the affinity for CO2 dropping and the oxygenase reaction being relatively enhanced (Kubien and Sage, 2008). To achieve acceptable high rates of photosynthetic CO2 fixation, typical C3 species invest up to 30% of soluble protein and some 25% of leaf nitrogen into Rubisco protein. Evolution of the CCM in C4 plants effectively circumvented a number of the inefficiencies, creating the present-day impetus for attempting to introduce C4 CCMs into important C3 crops such as rice (Hibberd et al., 2008). However, while the C4 CCM is one approach to elevating CO2 around Rubisco, drawing from our knowledge of single-cell CCMs in cyanobacteria (Price et al., 2008), there are also opportunities to elevate CO2 around Rubisco at the individual leaf chloroplast level. These prospects are expanded upon below, but in brief we consider two scenarios. The first, and simplest, approach is to consider the transplantation of cyanobacterial bicarbonate transporters to the C3 chloroplasts to provide marginal but significant improvement in photosynthetic performance. The second, more elaborate, longer term objective would be to engineer a more functional cyanobacterial CCM in the chloroplast. Cyanobacteria have evolved an extremely efficient CCM (Fig. 1; see below), being able to concentrate CO2 around Rubisco by a factor of up to 1,000-fold. As a result, cyanobacterial CO2 fixation has been able to retain a Rubisco with a relatively high carboxylation rate, although lower selectivity between CO2 and O2, compared with the Rubisco in C3 plants (Badger et al., 1998). Cyanobacterial cells also have high nitrogen use efficiency, as less nitrogen is devoted to Rubisco than in a C3 plant (Badger et al., 1998). In addition, Rubisco within a cyanobacterium operates at near CO2 saturation due to the action of the CCM, such that wasteful photorespiration is largely eliminated. The cyanobacterial CCM utilizes up to five uptake systems for DIC and the polyhedral microcompartments known as carboxysomes, which contain the cell’s complement of Rubisco and act as a localized site for the elevation of CO2 around Rubisco. A key operational feature is that all uptake systems deliver HCO3 − to the general cytoplasm and that this is kept in a state of dynamic disequilibrium favoring HCO3 −owing to the recycling of internally generated CO2 through the CO2 pumps and the absence of CA in the general cytoplasm. A specific, low level of CA activity is only present in the carboxysomes. In general, BicA, SbtA, BCT1, and NDH-I3 uptake systems are only induced under DIC-limiting conditions (e.g. liquid in near equilibrium with air CO2 levels or less). The carboxysomes are typically 90 to 200 nm in diameter (enlarged in this schematic), and a cell may possess five to 15 carboxysomes. By comparison, typical unicellular cyanobacteria are up to 3 μm in length, and typical C3 chloroplasts are up to 50 μm in diameter. Given that an early cyanobacterial progenitor is considered to have become the original endosymbiont for chloroplast evolution in algae and land plants, the question arises as to why present-day land plants lack any apparent chloroplast-based CCM. Cyanobacterial progenitors first appeared some 2.7 billion years ago (Buick, 1992), but it is almost certain that cyanobacteria have been subjected to periods of rapid evolutionary change throughout this period. In particular, the marked drop in CO2 levels, and the rise in O2 levels, that occurred around 400 to 350 million years ago (Berner, 1990) represents a likely trigger that forced the evolution of adaptations to cope with photorespiration and low-efficiency CO2 fixation (Fig. 2). The poor availability of CO2 in water, where diffusion is 104 times slower than in air and where large unstirred layers can exist, probably provided additional evolutionary pressure. In addition, with a pKa around 6.4, CO2 is a rarer species at alkaline pH, whereas HCO3 − is considerably more abundant in many aquatic environments. Evolutionary adaptations to deal with these combined pressures would have included transporters for the active uptake of dissolved inorganic carbon species (DIC; CO2 and HCO3 −), the subsequent localized elevation of CO2 around Rubisco, and the partitioning of Rubisco into microcompartments known as carboxysomes (Badger et al., 2002; Price et al., 2008; see below). This may have also been the stage when microalgae developed CCMs. If, as seems likely, cyanobacteria did not evolve fully functional CCMs until 350 million years ago, then this is well after the first terrestrial plants are thought to have evolved from eukaryotic algae at around 450 million years ago (Kenrick and Crane, 1997) and long after the original endosymbiotic event that gave rise to microalgae at around 1.5 billion years ago (Dyall et al., 2004). This probably explains why present-day crop plants lack any form of chloroplast-based CCM derived from cyanobacterial or microalgal ancestors. A timeline indicating that CCMs possibly arose in cyanobacteria and microalgae at around 400 to 350 million years ago, well after the evolution of early land plants. The cyanobacterial CCM functions to actively transport and accumulate DIC into the cell, where the accumulated HCO3 − pool is utilized to generate elevated CO2 levels around Rubisco (Badger et al., 2002; Price et al., 2008). Rubisco is encapsulated in unique microcompartments known as carboxysomes that are typically 90 to 200 nm in diameter. The functional importance of these proteinaceous, icosahedral bodies that are composed of 20 equilateral triangular sides is that they act as the site of CO2 elevation within the cell, with the supply rate of CO2 from accumulated HCO3 − being catalyzed by a carboxysome-located carbonic anhydrase (CA). The carboxysome shell in these cyanobacteria is composed of just six to eight proteins (Price et al., 2008), and the average unicellular cyanobacterial cells would normally possess five to 15 carboxysomes per cell. The key to the efficiency of any CCM revolves around the ability to minimize the loss of CO2 from the elevation zone. In model cyanobacteria, this is accomplished by a combination of (1) the accumulation of the ionic form of DIC, which is less membrane permeable than CO2, (2) the complete elimination of CA activity from the general cytosol to help reduce CO2 leakage out of the cell, (3) the special properties of the carboxysome protein shell acting to retard CO2 leakage, and (4) the action of the CO2 pumps in recycling CO2 leakage from the carboxysome back into the HCO3 − pool (Maeda et al., 2002; Price et al., 2008). The localization of CA, which catalyzes the reversible hydration and dehydration of CO2 and HCO3 −is a key element of cyanobacterial CCMs. The absence of CA in the cytosol, and the action of the directional CO2 uptake systems that convert CO2 to HCO3 − at the thylakoid membrane, allow the cell to accumulate HCO3 −keep it out of rapid chemical equilibrium with CO2. This is very effective in minimizing the concentration of the diffusible CO2 molecule owing to the slow dehydration of HCO3 − in the absence of CA (Walker et al., 1980). The importance of accumulating HCO3 − in the cytosol, and maintaining an internal HCO3 − pool out of chemical equilibrium, was shown by an experiment where human CA was expressed in the cytoplasm of a model cyanobacterium, Synechococcus elongatus PCC7942. The ectopic expression caused complete dissipation of the accumulated HCO3 − pool due to the CA-mediated equilibration between CO2 and HCO3 −which in turn led to increased CO2 diffusion out of the cell (Price and Badger, 1989). This is very different from the situation in C3 chloroplasts, where CA is highly abundant in the stroma in order to maximize the diffusion of CO2 across the envelope and throughout the chloroplast (Badger and Price, 1994). Five distinct transport systems for DIC uptake have been identified in cyanobacteria (Fig. 1; Table I; for more details and related references, see Price et al., 2008). (1) BCT1, which is inducible under DIC limitation and is a high-affinity HCO3 − transporter (uniporter) belonging to the traffic ATPase family. (2) SbtA, an inducible, high-affinity Na+-dependent HCO3 − transporter (Price et al., 2004; Shibata et al., 2002) that apparently acts as a Na+/HCO3 − symporter with relatively low flux rate. (3) BicA, a low-affinity, high-flux, Na+-dependent HCO3 − transporter belonging to the widespread SulP family and related to the human SLC26 family of anion transporters (Price et al., 2004); BicA is a probable Na+/HCO3 − symporter. (4) NDH-I4, a constitutive CO2 uptake system based on a specialized NADPH dehydrogenase (NDH-I) complex; this system uses NADPH as an electron donor to drive the conversion of CO2 to HCO3 − during the uptake step (Price et al., 2002). Each complex is composed of 10 core subunits that are common to the respiratory NDH-I complex and three specialized subunits required for CO2 uptake. Interestingly, NDH-I-type CO2 uptake systems appear to be located on the thylakoid membranes, where they use CO2 diffusing from outside the cell or arising from leakage from the carboxysomes as a substrate for directional conversion to HCO3 −. (5) NDH-I3, a second CO2 uptake system based on a modified NDH-I complex that is inducible under DIC limitation and is of higher uptake affinity than NDH-I4, located on the thylakoid membranes in Synechocystis PCC6803. With the objective of attaining a modest elevation of CO2 levels in the C3 chloroplast, the simplest approach would be to express a cyanobacterial HCO3 − transporter on the inner envelope of the C3 chloroplast (Fig. 3). Single-subunit HCO3 − transporters such as BicA and SbtA are the most obvious initial candidates. However, within technical restraints, the transfer of multisubunit transporters such as the BCT1 HCO3 − transporter (four genes) is also possible. Additionally, the use of HCO3 − transporters from microalgae such as Chlamydomonas can also be considered as viable candidates (Duanmu et al., 2009). From a technical viewpoint, the addition of DIC transporters mentioned above would be dependent on host genome transformation techniques using Agrobacterium tumefaciens, which are generally available for a range of important crop species. Chloroplast transformation techniques would not be required for this approach, and this is especially important because chloroplast transformation in crop species in not yet available. As can be seen from the associated modeling presented in this report (Fig. 4), the approach of installing BicA and/or SbtA transporters into the chloroplast inner envelope could achieve a 5% to 15% improvement in photosynthetic CO2 fixation rates at constant substomatal CO2 levels (see below). Schematic representations illustrating the concepts of adding a cyanobacterial HCO3 − transporter to the chloroplast envelope of a notional C3 leaf chloroplast (A) and the longer term prospect of constructing a more fully functional cyanobacterial or microalgal CCM in the C3 chloroplast (B). The diagrams show CO2 moving from the intracellular airspace (IAS; substomatal cavity) of a mesophyll leaf cell through the cell wall to the cytoplasm (Cyt) before entering the chloroplast by CO2 diffusion or via entry through a HCO3 − transporter. The hexagonal structure represents the icosahedral carboxysomes that would contain the full complement of Rubisco in the chloroplast, with a specific CA partitioned to this compartment and stromal CA removed. Linkages between the carbon reduction in the chloroplast and photorespiration involving peroxisomes (P) and mitochondria (M) are also shown. A, Modeled net CO2 assimilation rate (A) as a function of intercellular CO2 partial pressure (Ci), with three different options for bicarbonate transport at the chloroplast envelope: (1) a BicA transporter with a maximum activity of 30 μmol m−2 s−1 and k 1/2 for HCO3 − = 90 μm (approximately 140 μbar CO2 at pH = 7.4); (2) a SbtA transporter with a maximum activity of 15 μmol m−2 s−1 and k 1/2 for HCO3 − = 5 μm (approximately 11 μbar CO2 at pH = 7.4); (3) with both transporters. Only light-saturated CO2 assimilation is considered, and maximal Rubisco activity was = μmol m−2 The diffusive across the cell wall and across the chloroplast and m−2 s−1 This results in a of m−2 s−1 CO2 assimilation rates are compared with C3 photosynthesis with the maximal Rubisco and model are in the Table at higher as it not consider RuBP The between chloroplast and intercellular CO2 − as a function of for the in per net CO2 assimilation rate for the in the transport we per HCO3 − by BicA and per HCO3 − by is that a CO2 diffusion or drawdown between the CO2 level in the substomatal of the leaf and the level of CO2 in chloroplast with the of this at high (Evans and von Caemmerer, 1996). is important to that in the first the objective of adding a HCO3 − would be to the of this CO2 drawdown at the chloroplast and not to elevate it above the CO2 This the of wasteful CO2 a situation is very to the of a C4 into C3 cells et al., which has been for transplantation into a typical C3 chloroplast Caemmerer, and to be of the CO2 level within the chloroplast. specific modeling on the of BicA into a chloroplast is shown in and for the addition of HCO3 − pumps to there is need to CA levels in the chloroplast, since CA is to rapid equilibrium between accumulated HCO3 within the In of active HCO3 − uptake across the chloroplast the question arises as to a Na+-dependent HCO3 − transporter could function in a chloroplast. that at μm HCO3 − is present in the cytosol of a leaf cell in ambient air (Evans and von Caemmerer, and this to be by CA The uptake of SbtA flux and BicA flux for HCO3 − in cyanobacteria are 5 to 15 μm and 90 to et al., 2002; Price et al., and would that transporter would well above its K additional to the question of and the of across the chloroplast SbtA and BicA for activity in the form of a for et al., 2002; Price et al., 2004). The leaf cytosol to 3 et al., and have that the chloroplast envelope transporters and that are to cyanobacterial et al., there are prospects that the chloroplast and an As a the transfer of a cyanobacterial a from the C4 could also be considered this to be at the expense of any from the of transporters in the envelope et al., of and pH in the chloroplast by elevating HCO3 − levels by up to or by as as to would be to be et al., in SbtA or BicA in C3 chloroplasts would the to the chloroplast envelope and these transporters need to be In the of we that SbtA and BicA can be to the for known proteins such that details on are not have the membrane structure of BicA and SbtA as an initial step in the most likely in these transporters et al., be that there has been one to a cyanobacterial DIC in and in a improvement in use efficiency the of CO2 assimilation rate to in plants under in particular, a drop in the CO2 was et al., The of this is not yet since it is known that not for a DIC transporter et al., 2002; Price et al., 2008), and its in cyanobacteria is modeling of the of the first step of adding one or two cyanobacterial HCO3 − transporters to a C3 chloroplast is based on to consider the addition of a CO2 of single-cell C4 Caemmerer, von and and in the shown in are in the and Table Much of the of the of of single-cell C4 photosynthesis into a C3 leaf to the of bicarbonate transporters Caemmerer, The key from the modeling is that the addition of HCO3 − BicA or SbtA, can to an increase in the rate of light-saturated CO2 assimilation at ambient and low intercellular CO2 partial pressures The of the increase be very dependent on the kinetic properties of the transporters and the to CO2 diffusion of the chloroplast envelope Caemmerer, The of a transporter chloroplast CO2 partial pressures above at low in a CO2 The addition of the high-affinity SbtA transporter is more effective at the than the BicA transporter because of its lower K and of both can be more effective higher levels, transporters to reduce the drawdown in CO2 between intercellular CO2 and the chloroplast (Fig. a constant level of a with the of both HCO3 − uptake could an assimilation rate than 15% higher than of the of C4 photosynthesis is the of two per CO2 in the C4 which the of a C4 into C3 Caemmerer, although there are of a number of single-cell C4 species that have the limitations inherent in C3 et al., 2004). is likely that bicarbonate transport is less the of HCO3 − uptake in cyanobacteria et al., and the likely we a required per HCO3 − by BicA and per HCO3 − for this the per net CO2 assimilation rate with during C3 photosynthesis as the cost of photorespiration (Fig. The of bicarbonate transporters the cost at low that normally during C3 photosynthesis and above the C3 at higher (Fig. have only considered the for light-saturated photosynthesis crops, the be under these conditions, which the of bicarbonate transporters a leaf photosynthetic rates also has the of leaf on transporters provide the of a low and the of a plant with HCO3 − could to be less while the rate of in less loss of from the leaf The SbtA transporter could be of under conditions more effectively than BicA, and the addition of both transporters is likely to be SbtA or SbtA BicA can be into C3 plants, such species be able to better of high under high and A longer term involving the technical of the of could be to a more form of the cyanobacterial CCM in the chloroplast (Fig. 3). This could the transfer of one or two active HCO3 − transporters to the inner envelope membrane combined with the transfer of a CO2 uptake system to the thylakoid membranes and a Rubisco such as the carboxysomes. The C3 chloroplast would also need to the cyanobacterial cytosol, to a HCO3 where the HCO3 − pool is in a state of slow chemical To it would be to C3 Rubisco into effective carboxysome and an effective of the highly abundant CA that HCO3 − accumulation can be it has been to up to of CA activity in by (Price et al., 1994). a complete of CA from the stroma would be more for the important CA in carboxysomes. of the most significant to the of the envelope to CO2 with a range of available et al., 2008). to a in CO2 a be to reduce levels in the envelope by since lower would in CO2 leakage et al., of carboxysome and function has in to the where the of a carboxysome in the chloroplast is This has been by in the of some key of the shell et al., and our to proteins required as key and proteins et al., The feature of the shell proteins is an ability to et al., 2008). This with some could the of functional carboxysomes within the chloroplast. The longer term objective of a more form of the cyanobacterial CCM into the chloroplast may provide photosynthetic than the of bicarbonate transporters that the addition of cyanobacterial HCO3 − pumps at the chloroplast envelope of a typical C3 plant could provide a significant to the photosynthetic of leaf photosynthesis, as increased assimilation rate or as A is to BicA and SbtA transporters to the chloroplast of a model C3 plant and to our of and for cyanobacterial HCO3 − transporters. However, it also be that such as the of Rubisco or for RuBP or would provide to crop as in in this The are available in the of this in of in Table in the model